Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
This office action is responsive to the amendment filed on 9/16/2026. As directed by the amendment, the status of the claim(s) are:
Claim(s) 4-5, 8, 18-19 has/have been amended;
Claim(s) 1-20 is/are presently pending.
Response to Arguments
Applicant argues on p. 7 of remarks that Woods does not teach “identifying a region of interest surrounding at least one lead”. After review, this is not persuasive because Woods teaches measuring/imaging and analyzing voxels or regions within brain due to stimulation by the electrodes, these voxels surround the electrode because the dictionary definition of “surround” is “to enclose on all sides”. [0042] “the electro-stimulation device 170 generates an electric field within the volume…over a plurality of volume elements or voxels 122 that are defined within a frame of reference of the electro-stimulation device”; [0054] “Model solutions generated in all electrode locations provided 71 unique electric field distributions to serve as the lead field for each head model…the net electric field for any electrode pair was computed”; [0072] “a final model was trained on all fourteen current density maps”; [0074] “top ten ranked regions of interest (ROIs) based on the mean percent contribution per voxel within each region…the top ten ROIs…were largely located in the frontal region underneath and between the electrodes”; [0081] “electrode orientation and location…can also dictate the shape and location of stimulation contact area, altering the pattern of current flow within the brain…monitoring and increasing electrode placement accuracy…may improve individual treatment response”. See also Fig. 2A, 200 “Define a set of voxels in a fixed reference frame of an electro-stimulation device including a plurality of electrodes positioned on a head of a subject”. As can be seen in Fig. 4A-4B, where electrodes are placed on top of head, Fig. 8A-D, Fig. 9A then takes the voxels of the brain in which the region surrounds the electrodes; compare especially Fig. 4A to Fig. 9A.
Applicant argues on p. 7 that Woods does not show “generating a database of previously tested stimulation settings and their clinical effect”. After review, this is not persuasive. Woods Fig. 2A depicts a method in which stimulation is applied and whether a cognitive effect in the subject is achieved is measured this data is stored to be used for machine learning in order to best predict/suggest treatment for a new patient in Fig. 2B and Fig. 11A-F. Stimulation setting are adjusted, see paragraph [0050] “specific values for various parameters of the EBS (e.g. current intensity value, number of stimulation sessions, period time over which the sessions were conducted, conductivity values of the electrode and gel layers, etc.) these parameter values are not limiting and in other example embodiments, one or more of these parameter values can be adjusted or varied”.
Applicant argues on p. 8 that Woods does not show trained machine-learning model “configured to classify the stimulation parameter value for the stimulation setting based at least in part on raw imaging data”. As discussed above, Woods teaches that stimulation settings can be adjusted and whether there is a clinical effect is analyzed. Fig. 2B and Fig. 11A-F. Stimulation setting are adjusted, see paragraph [0050] “specific values for various parameters of the EBS (e.g. current intensity value, number of stimulation sessions, period time over which the sessions were conducted, conductivity values of the electrode and gel layers, etc.) these parameter values are not limiting and in other example embodiments, one or more of these parameter values can be adjusted or varied”. The analysis is performed by machine learning, see Fig. 11 AI discriminator, [0102] “Artificial intelligence” and [0033] “machine learning” with the outcome of suggesting stimulation parameters to obtain the best therapeutic effect (Fig. 11; [0100] “The method disclosed herein is the provided to utilize MRI-derived computational models, machine learning, and dose optimization to maximize the likelihood of treatment response”).
As for “raw imaging data”, Woods teaches using imaging, specifically MRI to analyze the voxels and regions of interest due to electrical stimulation (Fig. 1A; Fig. 6; Fig. 11). The claim element is “based at least in part on raw imaging data” and so Woods meets this limitation just by virtue of utilizing MRI imaging, regardless of how much “raw” data is used in the analysis of voxels/region of interest of brain.
Applicant argues on p. 9 that Woods does not teach the recited “similarity metric between the patient of interest and a plurality of patients according to the region of interest”. After review, this is not persuasive because Woods is teaching taking a patient and comparing to previously tested subjects and predicting whether patient is a responder or nonresponder and which stimulation parameters would best give a desired therapeutic effect. [0100] “The method disclosed herein is the provided to utilize MRI-derived computational models, machine learning, and dose optimization to maximize the likelihood of treatment response”). Fig. 2B depicts the steps for when a patient is compared to previous responder/nonresponder type of Fig. 2A. [0087] “this comparison is used to assess whether the new subject is in the first or second group of subjects. In these embodiments, if the new subject is in the second group of subjects, step 257 includes one or more steps to adjust parameter values of the electro-stimulation device 170 to improve a cognitive response of the new subject…if the new subject is in the second group, then one or more parameter values of the electro-stimulation device 170 are adjusted”). The assessment of whether new subject is in the first or second group reads on the recited “similarity metric”.
Applicant argues on p. 9 that Woods does not teach the further requirements to the similarity metric of claims 14-15, 17. After review, this is not persuasive. As discussed above, Woods teaches taking a new subject and determining if they are within a responder or nonresponder group in order to predict and suggest optimal stimulation parameters. Woods [0095] specifically teaches “The likelihood (l) of a new subject belonging to the responders’ current distribution was calculated…used as a fitness metric to optimize tDCS parameters…After the optimization process, current density volumes…were passed back through the original SVM model for re-classification as responders or non-responders”.
Applicant argues on p. 9 that Woods does not teach claims 9 and 19’s hierarchical organization. After review this is not persuasive. Woods teaches in [0061] “current models were passed through an SVM algorithm to characterize cranial EBS (e.g. tDCS) current components (e.g. intensity and direction) that induced working memory improvements…In an embodiment,…feature selection was performed on the training data to further reduce the number of trained features. One popular method of feature selection is to filter the features via voxel wise t-tests between classes to select current elements with a significant group-level difference (p<0.01) as features for the subsequent prediction step”. [0068] “A secondary set of analyses was aimed at determining the characteristics of current with the voxels determined to be essential for treatment response.” All of this is to obtain [0100] “The method disclosed herein is the provided to utilize MRI-derived computational models, machine learning, and dose optimization to maximize the likelihood of treatment response”. Thus, Woods teaches using filtering in order to obtain the parameters that are most important or most likely to make a difference in treatment outcome and this reads on “hierarchal organization” of the instant claims because the most important parameters are found and utilized.
Applicant argues on p. 9-10 that Woods does not teach data-driven or anatomical phenotypes of claims 8, 13, 19. After review, this is not persuasive, Woods teaches grouping subjects into responder vs nonresponder and this is based on analysis of their tissue type which is a type of phenotype. See Fig. 8A-9; Fig. 11A-11F.
Applicant argues on p. 10 that Woods does not teach claim 7’s additional data types beyond raw imaging data. After review, this is not persuasive. Woods as cited teaches using result of cognitive test, whether subject is responder or nonresponder as part of analysis. This is a different data type than raw imaging data alone.
Applicant argues on p. 10 that rejection of claim 11 seems to have had a typo “-20”; Examiner apologizes for this error and has corrected it.
Applicant argues on p. 10 for confirmation of citation supporting the preamble of claim 1 due to typo. Examiner apologizes for this error and has corrected it.
The citations in the rejection(s) below have been updated in light of the discussion above. This does not constitute a new grounds of rejection; MPEP 1207.03 II Citing a different portion of a reference to elaborate upon that which has been cited previously.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-9, 11-20 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Woods (US 20230293899 A1; 9/21/2023).
Regarding claim 1, Woods teaches a method for machine-learning based automated neurostimulation programming ([0004]; [0102]; [0057]-[0059]), the method comprising:
generating a database of previously tested stimulation settings and their clinical effect (Fig. 2A, 214; Fig. 11A-11F; [0050] “specific values for various parameters of the EBS (e.g. current intensity value, number of stimulation sessions, period time over which the sessions were conducted, conductivity values of the electrode and gel layers, etc.) these parameter values are not limiting and in other example embodiments, one or more of these parameter values can be adjusted or varied”);
identifying a region of interest surrounding at least one lead (Fig. 2A, 200; Fig. 11A-11F; Fig. 4A-4B, where electrodes are placed on top of head, Fig. 8A-D, Fig. 9A then takes the voxels of the brain in which the region surrounds the electrodes; compare especially Fig. 4A to Fig. 9A);
determining a stimulation setting corresponding to one or more specific clinical effects associated with the region of interest, the one or more specific clinical effects including at least one beneficial clinical effect or at least one detrimental clinical effect (Fig. 2A, 212; Fig. 11A-11F; [0057]-[0059]; [0050] “specific values for various parameters of the EBS (e.g. current intensity value, number of stimulation sessions, period time over which the sessions were conducted, conductivity values of the electrode and gel layers, etc.) these parameter values are not limiting and in other example embodiments, one or more of these parameter values can be adjusted or varied”);
employing a trained machine-learning model to identify a stimulation parameter value for a stimulation setting to set in a patient of interest based on the one or more specific clinical effects, the trained machine-learning model configured to classify the stimulation parameter value for the stimulation setting based at least in part on raw imaging data (Fig. 2B; Fig. 11A-11F; [0060]; [0102]; Fig. 1A; Fig. 6); and
generating an output providing the stimulation parameter value for the stimulation setting for the patient of interest (Fig. 2B; Fig. 11A-11F; [0060]; [0102]; [0100] “The method disclosed herein is the provided to utilize MRI-derived computational models, machine learning, and dose optimization to maximize the likelihood of treatment response”).
Claim 16 is rejected under substantially the same basis as claim 1 above.
Claim 20 is rejected under substantially the same basis as claim 1 above.
Regarding claim 2, Woods teaches determining a similarity metric between the patient of interest and a plurality of patients according to the region of interest (Fig. 12A-12F; [0060]; [0097]; [0098]; [0099] “predicted treatment response”; [0100]).
Regarding claim 3, Woods teaches using a single similarity metric or multiple similarity metrics (Fig. 12A-12F; [0060]; [0097]; [0098]; [0099] “predicted treatment response”); and
clustering stimulation settings from similar patients identified among the plurality of patients according to the region of interest (Fig. 11A-11F; Fig. 12A-12F; [0060]; [0097]; [0098]; [0099] “predicted treatment response”; [0100]; [0095] “The likelihood (l) of a new subject belonging to the responders’ current distribution was calculated…used as a fitness metric to optimize tDCS parameters…After the optimization process, current density volumes…were passed back through the original SVM model for re-classification as responders or non-responders”).
Claim 17 is rejected under substantially the same basis as claims 2-3 above.
Regarding claim 4, Woods teaches suggesting or avoiding the stimulation settings based on cluster features identified among the plurality of patients according to the region of interest (Fig. 11A-11F; Fig. 12A-12F; [0097]; [0098]; [0099] “predicted treatment response”; [0100]).
Regarding claim 5, Woods teaches identify one or more commonalities between the patient of interest and the plurality of patients in the region of interest around the lead (Fig. 11A-11F; Fig. 12A-12F; [0060]; [0097]; [0098]; [0099] “predicted treatment response”; [0100]).
Regarding claim 6, Woods teaches employing the trained machine-learning model to identify the one or more commonalities based at least in part on additional data types beyond raw imaging data (Fig. 2A, 212, 214; [0069] “cognitive test”; [0077] “Introducing additional neuroimaging modalities and clinical data into the model may further enhance performance and predictive value”; claim 12).
Regarding claim 7, Woods teaches wherein the additional data types beyond raw imaging data include at least one of disease state, comorbidities, patient states, or clinician evaluation (Fig. 2A, 212, 214; [0069] “cognitive test”; [0077] “Introducing additional neuroimaging modalities and clinical data into the model may further enhance performance and predictive value”; claim 12).
Claim 18 is rejected under substantially the same basis as claims 5-7 above.
Regarding claim 8, Woods teaches identifying the stimulation parameter value for the stimulation setting to set in the patient of interest based on data-driven phenotypes (Fig. 7A; non-responders vs responders; Fig. 2A, 214; Fig. 12A-12F; [0060]; [0100]-[0102]; Fig. 8A-9; Fig. 11A-11F).
Regarding claim 9, Woods teaches identifying the stimulation parameter value for the stimulation setting to set in the patient of interest based on a hierarchical organization ([0061] “feature selection is to filter the features…significant group-level difference”; [0102]; [0068] “A secondary set of analyses was aimed at determining the characteristics of current with the voxels determined to be essential for treatment response.”).
Claim 19 is rejected under substantially the same basis as claims 8-9 above.
Regarding 11, Woods teaches employing the trained machine-learning model to confirm that the model-identified stimulation parameter value for the stimulation setting to set in the patient of interest based on the one or more specific clinical effects matches known target settings for a patient (Fig. 11A-11F; Fig. 12A-12F; [0060] “The inventors of the present invention recognized that many parameter values of the electric field from step 210 do not considerably vary between the first group and second group of subjects, whereas other parameter values of the electric field do considerably vary between the first group and second group of subjects. The inventors of the present invention recognized that if those parameter values of the electric field that distinguish the first group of subjects from the second group of subjects were determined, they could be used to efficiently predict whether a new subject will experience cognitive function improvement based on parameter values of the electric field in the new subject.”; [0102] “It was shown that the optimal stimulation parameters reduce between-subject variability, elevated current intensity within targeted brain regions, and closely resembled the current distributions observed in treatment responders.”).
Regarding claim 12, Woods teaches wherein the confirmation is performed through use of at least one of computational modeling ([0099] “Computational models…predicted treatment response with over 86% accuracy”), imaging, clinical evaluation, symptomology scoring (Fig. 2A-2B; Fig. 11A-11F; Fig. 12A-12F; [0060]; [0102]), or invasive testing.
Regarding claim 13, Woods teaches employing the trained machine-learning model to determine target similarity metrics and thresholds for grouping patients by anatomical phenotypes (Fig. 2A-2B; Fig. 7A-7D; Fig. 10A-10D; Fig. 11A-11F; Fig. 12A-12F; [0072]; [0099]; Fig. 8A-9; Fig. 11A-11F).
Regarding claim 14, Woods teaches wherein grouping patients by the anatomical phenotypes includes using different similarity metric correlations, varying the thresholds to evaluate accuracy of predictions, identifying the thresholds that maximize accuracy, compute different similarity metrics on various image features, or selecting the various image features that produce distinct patient groupings (Fig. 2A-2B; Fig. 7A-7D; Fig. 10A-10D; Fig. 11A-11F; Fig. 12A-12F; [0072]; [0099]; [0095] “The likelihood (l) of a new subject belonging to the responders’ current distribution was calculated…used as a fitness metric to optimize tDCS parameters…After the optimization process, current density volumes…were passed back through the original SVM model for re-classification as responders or non-responders”).
Regarding claim 15, Woods teaches validating, by the trained machine-learning models, a particular similarity metric and threshold by testing prediction accuracy on a new patient ([0077] “we used two-level nested cross-validation to increase the number independent test samples and used 10 permutations to assess the retest reliability of these models. Average accuracy and confidence intervals across these permutations were used to estimate model performance on novel datasets.”; [0099] “Computational models of current intensity and direction predicted treatment response with over 86% accuracy”).
Allowable Subject Matter
Claim(s) 10 is/are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The prior art of record does not disclose or fairly suggest either singly or in combination the claimed invention of claim 10 when taken as a whole, comprising, in addition to the other recited claim elements, wherein the hierarchical organization includes a first level providing a best contact, a second level providing a best contact fraction, and a third level providing a best amplitude.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jonathan T Kuo whose telephone number is (408)918-7534. The examiner can normally be reached M-F 10 a.m. - 6 p.m. PT.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Niketa Patel can be reached at 571-272-4156. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JONATHAN T KUO/ Primary Examiner, Art Unit 3792
